1. Field of the Invention
The present invention relates to an exhaust gas flow measuring equipment for measuring the rate at which gas is discharged from internal combustion engines and to processes for calibrating the sensitivity of a trace gas flow meter.
2. Description of the Prior Art
In order to carry out transient characterization of gas discharged from internal combustion engines (hereinafter call "exhaust gas"), the exhaust gas flow rate must be measured in real time. A trace method is one technique used to continuously measure the exhaust gas flow rate. This trace method introduces inert gas, for example, helium gas, which does not react with the components in the exhaust gas to the exhaust passage linked to the internal combustion engine. The method then measures the helium gas concentration with a trace gas analyzer connected to the gas sampling passage connected to the exhaust passage. The exhaust gas flow rate is then determined in real time by dividing the introducing rate of the helium gas by the concentration of the helium gas.
Examples of equipment for measuring the exhaust gas flow rate operating on the above-mentioned conventional measuring principle include that disclosed in Japanese non-examined Patent Publication No. Hei 8-15253. FIG. 22 schematically shows conventional engine exhaust gas flow rate measuring equipment disclosed in this patent publication. Numeral 71 designates an engine, numeral 72, a compressed gas cylinder for introducing helium gas as inert gas into this engine 71, and numeral 73, a pressure reducing valve. Numeral 74 designates an exhaust passage linked to the engine 71. Numeral 75 designates a gas sampling passage branched and connected to the exhaust passage 74 at the upstream side, which is equipped with a filter 76 and a suction pump 77, and joined and connected to the exhaust passage 74 on the downstream side. Numeral 78 designates a trace gas analyzer connected to the gas sampling passage 75 via a connecting member 79.
The trace gas analyzer 78 may be a quadruple mass spectrometer, sector field mass spectrometer, or similar device, but since these analyzers have a high vacuum inside, microleakage orifice or variable leak valve (VLV) is used for the connecting member 79 in connecting to the gas sampling passage 75 to which the filter 76, the suction pump 77, etc. are installed.
In measuring the exhaust gas flow rate, for example, helium gas must be introduced as a trace gas into the exhaust pipe 74a linked to the engine 71. But conventionally, as shown in FIG. 23, a pipe 75 comprising of, for example, tetrafluorethylene resin which is strong to, for example, exhaust gas G and can withstand comparatively high temperature, is inserted and connected to cross nearly at right angles to the direction in which the exhaust gas G flows. Helium gas TG is introduced as trace gas to the exhaust pipe 74a in which exhaust gas G flows via this pipe 75. However, in the above-mentioned configuration a number of problems exist as described below, and the measurement accuracy of the exhaust gas flow rate is not always satisfactory.
As the microleakage orifice of VLV has a large inside dead volume, when a plurality of other gas analyzers are connected to the gas sampling passage 75 and exhaust gas components such as CO, CO.sub.2, NO.sub.x, HC, etc. are analyzed with these gas analyzers, lag time is generated in the trace gas analyzer 78 and the gas analyzer, and the output timing must be adjusted in both analyzers.
As the inside of the trace gas analyzer 78 is originally of high vacuum, the sensitivity varies in accordance with the gas component ratio in the exhaust gas. That is, when the trace gas analyzer 78 has its temperature adjusted to a specified level, helium gas is introduced while being mixed in the exhaust gas at a specified concentration via the connecting member 79. In the continuous measurement of exhaust gas discharged from the internal combustion engine such as automobile engines, if the exhaust gas component to be measured suddenly changes, the difference is generated in pressure inside the trace gas analyzer 78 due to the difference of viscosity depending on this exhaust gas component. If the pressure, volume, and temperature inside the trace gas analyzer 78 are denoted by P, V (constant), and T (constant), then the equation PV=nRT (n: molecular number of helium gas and R: constant) holds for the helium gas. But when the pressure change .DELTA.P is, for example, positive, since P is proportional to n in the above equation, the introducing amount of helium gas increases, and the reading of the helium gas in the exhaust gas becomes higher than the actual value. Also, the low exhaust gas flow rate is indicated. On the contrary, if the pressure variation is negative, the helium introducing volume decreases in proportion to this variation, and then the reading of helium gas in the exhaust gas becomes lower than the actual value. And for exhaust gas flow rate, a higher value is obtained.
In FIG. 23, while the inside diameter of exhaust pipe 74a is as large as 100 mm, that of the pipe 75 for introducing helium gas is about 4 mm. As pipe 75 is inserted in such a manner to simply cross at right angles with the flowing direction of exhaust gas with respect to the exhaust gas 74a, mixing of the exhaust gas G from the engine with helium gas TG does not always take place satisfactorily. Consequently, errors occur in the helium gas concentration measurement results by the trace gas analyzer 78, and there has been an inconvenience in that the measurement accuracy of the exhaust gas flow rate is not always satisfactory.
On the other hand, with respect to the sensitivity calibration method, conventionally pure nitrogen gas (N.sub.2) is used for zero gas. At the same time, a mixture of several tens to several thousands ppm of helium gas is added with pure N.sub.2 as base for span gas which is used for zero calibration and span calibration of the trace gas analyzer. Problems analogous to those described above exist with this configuration.
As no consideration is given to carbon dioxide (CO.sub.2) contained in a large quantity next to N.sub.2 in the exhaust gas and as calibration was carried out, the sensitivity change and the desired sensitivity calibration are unable to be carried out. With regard to the calculation of continuous mass emission rate from a car, the measured flow rate and a gas concentration for each gaseous constituents must be multiplied. In the conventional way, since the flow rate is always measured as a whole, gaseous constituents and the gas concentration is typically dehumidified concentration, either the measured flow rate must be converted to dehumidified concentration or the gas concentration must be converted to pre-humidified concentration with mathematical way. The conversion generates additional source of error in getting mass emission due to the water vapor concentration has to be assumed on the perfect combustion in the engine.